Real-Time Fault Diagnosis for Gas Turbine Blade Based on Output-Hidden Feedback Elman Neural Network

doi:10.1007/s12204-018-2028-4

Abstract

Abstract: In order to remotely monitor and maintain large-scale complex equipment in real time, China Telecom plans to create a total solution that integrates remote data collection, transmission, storage, analysis and prediction. This solution can provide manufacturers with proactive, systematic, integrated operation and maintenance service, and the data analysis and health forecasting are the most important part. This paper conducts health management for the turbine blades. Elman neural network, and improved Elman neural network, i.e., outputhidden feedback (OHF) Elman neural network are studied as the main research methods. The results verify the applicability of OHF Elman neural network.

Key words: gas turbine, blade| health management | output-hidden feedback (OHF) Elman neural network

摘要： In order to remotely monitor and maintain large-scale complex equipment in real time, China Telecom plans to create a total solution that integrates remote data collection, transmission, storage, analysis and prediction. This solution can provide manufacturers with proactive, systematic, integrated operation and maintenance service, and the data analysis and health forecasting are the most important part. This paper conducts health management for the turbine blades. Elman neural network, and improved Elman neural network, i.e., outputhidden feedback (OHF) Elman neural network are studied as the main research methods. The results verify the applicability of OHF Elman neural network.

关键词: gas turbine, blade| health management | output-hidden feedback (OHF) Elman neural network

CLC Number:

TP 183
V 23

ZHUO Pengcheng (卓鹏程), ZHU Ying (朱颖), WU Wenxuan (邬雯喧), SHU Junqing (舒俊清), XIA Tangbin (夏唐斌). Real-Time Fault Diagnosis for Gas Turbine Blade Based on Output-Hidden Feedback Elman Neural Network[J]. Journal of Shanghai Jiao Tong University (Science), 2018, 23(Sup. 1): 95-102.

References 21

[1]	WANG J T, LIU H W, XU D. A research on thelarge-scale complex equipment spare parts demandprediction based on the BP neural network modeland Markov chain [C]//Proceedings of 2013 InternationalConference on Industrial Engineering and ManagementScience. [s.l.]: Springer, 2013: 774-779.
[2]	CAI B P, LIU H L, XIE M. A real-time fault diagnosismethodology of complex systems using object-orientedBayesian networks [J]. Mechanical Systems and SignalProcessing, 2016, 80: 31-44.
[3]	ESLAMI M, SHAYESTEH M, POURAHMADI M R.Optimal design of PID-based low-pass filter for gas turbineusing intelligent method [J]. IEEE Access, 2018,6: 15335-15345.
[4]	WAN A P, GU F, CHEN J H, et al. Prognostics ofgas turbine: A condition-based maintenance approachbased on multi-environmental time similarity [J]. MechanicalSystems and Signal Processing, 2018, 109:150-165.
[5]	JIA X D, JIN C, BUZZA M, et al. Wind turbine performancedegradation assessment based on a novel similaritymetric for machine performance curves [J]. RenewableEnergy, 2016, 99: 1191-1201.
[6]	FORBES G L, RANDALL R B. Estimation of turbineblade natural frequencies from casing pressureand vibration measurements [J]. Mechanical Systemsand Signal Processing, 2013, 36(2): 549-561.
[7]	PENNACCHI P, CHATTERTON S, BACHSCHMIDN, et al. A model to study the reduction of turbineblade vibration using the snubbing mechanism [J].Mechanical Systems and Signal Processing, 2011, 25:1260-1275.
[8]	LIU C, JIANG D X. Crack modeling of rotating bladeswith cracked hexahedral finite element method [J]. MechanicalSystems and Signal Processing, 2014, 46: 426-423.
[9]	COMPARE M, BELLANI L, ZIO E. Reliability modelof a component equipped with PHM capabilities [J].Reliability Engineering and System Safety, 2017, 168:4-11.
[10]	JARDINE A K S, LIN D M, BANJEVIC D. A reviewon machinery diagnostics and prognostics implementingcondition-based maintenance [J]. Mechanical Systemsand Signal Processing, 2006, 20: 1483-1510.
[11]	XIA T B, DONG Y F, XIAO L, et al. Recent advancesin prognostics and health management for advancedmanufacturing paradigms [J]. Reliability Engineeringand System Safety, 2018, 178: 255-268.
[12]	ZHONG J H,WONG P K, YANG Z X. Fault diagnosisof rotating machinery based on multiple probabilisticclassifiers [J]. Mechanical Systems and Signal Processing,2018, 108: 99-114.
[13]	LEE J, WU F J, ZHAO W Y, et al. Prognostics andhealth management design for rotary machinery systems:Reviews, methodology and applications [J]. MechanicalSystems and Signal Processing, 2014, 42: 314-334.
[14]	HANACHI H, MECHEFSKE C, LIU J, et al.Performance-based gas turbine health monitoring, diagnostics,and prognostics: A survey [J]. IEEE Transactionson Reliability, 2018, 67(3): 1340-1363.
[15]	HU Q P, XIE M, NG S H, et al. Robust recurrentneural network modeling for software fault detectionand correction prediction [J]. Reliability Engineeringand System Safety, 2007, 92: 332-340.
[16]	OH K Y, PARK J Y, LEE J S, et al. A novel methodand its field tests for monitoring and diagnosing bladehealth for wind turbines [J]. IEEE Transactions on Instrumentationand Measurement, 2015, 64(6): 1726-1733.
[17]	LI Y F, LI G X. YAN J. Fault diagnosis of wind turbineblades based on fuzzy theory [C]//InternationalConference on Control, Automation and Systems Engineering.Singapore: IEEE, 2011: 1-3.
[18]	GAO J R, WANG Y Q. The research on the methodsof diagnosing the steam turbine based on the Elmanneural network [J]. Journal of Software Engineeringand Applications, 2013, 6: 87-90.
[19]	SHI X H, LIANG Y C, LEE H P, et al. ImprovedElman networks and applications for controlling ultrasonicmotors [J]. Applied Artificial Intelligence, 2004,18: 603-629.
[20]	WANG Y N, ZHANG F J, CUI T, et al. Fault diagnosisfor manifold absolute pressure sensor (MAP) ofdiesel engine based on Elman neural network observer[J]. Chinese Journal of Mechanical Engineering, 2016,29(2): 386-395.
[21]	DRAKE P R, MILLER K A. Improved self-feedbackgain in the context layer of a modified Elman neuralnetwork [J]. Mathematical and Computer Modelling ofDynamical Systems, 2002, 8(3): 307-311.